| 摘要: |
| 在大数据时代，如何高效地处理海量数据以满足性能需求，是一个需要解决的重要问题.内存计算充分利用大容量内存进行数据处理，减少甚至避免I/O操作，因而极大地提高了海量数据处理的性能，同时也面临一系列有待解决的问题.首先，在分析内存计算技术特点的基础上对其进行了分类，并分别介绍了各类技术及系统的原理、研究现状及热点问题；其次，对内存计算的典型应用进行了分析；最后，从总体层面和应用层面对内存计算面临的挑战予以分析，并且对其发展前景做了展望. |
| 关键词: 内存计算 新型混合内存 分布式集群 图计算 大数据处理 |
| DOI：10.13328/j.cnki.jos.005103 |
| 分类号: |
| 基金项目:国家自然科学基金（91530324）；国家高科技研究发展计划（863）（2015AA01A301） |
|
| Survey on In-Memory Computing Technology |
|
LUO Le, LIU Yi, QIAN De-Pei
|
|
School of Computer Science and Engineering, BeiHang University, Beijing 100191, China
|
| Abstract: |
| In the era of big data, systems need to process massive data efficiently to meet performance requirements of applications. In-memory computing technology can improve performance of massive data processing significantly by utilizing memory and avoid I/O operations. However, the technology faces a series of challenges that need to be solved. This paper analyzes characteristics of in-memory computing technology, lays out its classification, and introduces principles, related works and hot topics of every category. In addition, several typical applications of in-memory technology are introduced. Finally, challenges and opportunities for in-memory computing are elaborated. |
| Key words: in-memory computing hybrid memory distributed system graph processing big data processing |
